Reasoning & Chain-of-Thought - Weekly Roundup

Fysiverse-3D-Vision is proposed, a unified vision-language-geometry framework for generative 3D scene reconstruction and executable asset construction from a single image that establishes a shared representation where spatial reasoning and geometric reconstruction mutually enhance each other, allowing object layouts to be inferred beyond the constraints of individual asset generators.

A hierarchical framework that couples high-level multimodal reasoning with low-level function-constrained execution for deployable precision medical robotics, and introduces a scalable multi-agent pipeline to generate skill-oriented chain-of-thought data for structured training.

It is found that expert selection is far more redundant than the field's operating points assume, and how much expert computation fine-grained MoEs can dispense with is revealed, and when dynamic allocation earns its complexity is established, informing both practical deployment and future pruning methods.

A benchmark study of six TSFMs evaluated on energy, traffic, and financial datasets reveals that while TSFMs outperform statistical methods and supervised models, they are subject to a fundamental trade-off between point accuracy and probabilistic reliability.

It is proved that LEMA transformers with chain of thought can simulate word-RAMs, as was recently shown for the less restrictive rightmost hard attention, and dictionary-based inference is implemented for LEMA transformers with dictionaries residing in main memory rather than VRAM.

Across three models, two reasoning regimes, and two domains, four findings hold without per-turn reasoning: displaying accurate state is unreliable, an unverified ledger the agent writes itself beats an accurate checklist it is shown, directives help in proportion to the model's obedience, and enforcement needs no obedience but is bounded by the correctness of its state.

OmniFysics-Nano-V2, a compact omni-modal model for physical-world perception and understanding, is presented and experiments show that the proposed data and training strategy improves physical-world understanding while preserving broad omni-modal competence.

This work builds Video-HopChain, a dataset of 22,550 multi-hop video questions over 13,378 videos, together with a held-out benchmark of 1,000 questions, which raises the mean over eight video understanding and reasoning benchmarks from 55.4 to 57.9 and introduces Confidence-Gated Exploration (CGE).

A progressive four-stage framework that addresses three clinically critical gaps in TCM prescription generation and substantially improves prescription quality over zero-shot baselines is proposed, with a 7B model (Mistral-7B) surpassing zero-shot GPT-5 on all three TCM evaluation metrics.

A Coarse-to-Fine Scene Expert Network (C2FXNet) that achieves unified detection through hierarchical scene guidance through dual-level guidance mechanism, and enables robust multi-scene detection without scene-specific training.

This work introduces ACLArena, a framework for comprehensively studying, analyzing, and evaluating Agent Continual Learning, and proposes a new ACL recipe that combines offline replay over high-quality trajectories with a routed network of multiple LoRA experts each specialized via RL, substantially improving the agent's ability to learn across multiple domains.

A training-free agentic framework built around specialist detectors, per-detector triage, and conflict-aware evidence arbitration is studied, indicating that the main challenge of open-world image forensics is not detecting manipulations, but calibrating trust in specialized forensic tools and arbitrating conflicting evidence.

Qwen-Audio-3.1-Realtime brings these requirements together through Think, Act, and Speak and Coordinate, evaluating audio reasoning, multilingual understanding, tool use, conversational behavior, full-duplex interaction, and safety.